Skip to main content
ARS Home » Southeast Area » Stoneville, Mississippi » Biological Control of Pests Research » Research » Publications at this Location » Publication #220228
Title: Feasibility of Using Template-Based and Object Based Automated Detection Methods for Quantifying Black and Hybrid Iimported Fire Ant (Solenopsis invicta Buren and S. invicta x richteri) Mound in Aerial Digital Imagery

Author
item Vogt, James
item WALLET, BRADLEY - AUTOMATED DECISIONS & CPS

Submitted to: The Rangeland Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/18/2008
Publication Date: 8/27/2008
Citation: Vogt, J.T., Wallet, B. 2008. Feasibility of Using Template-Based and Object Based Automated Detection Methods for Quantifying Black and Hybrid Iimported Fire Ant (Solenopsis invicta Buren and S. invicta x richteri) Mound in Aerial Digital Imagery. The Rangeland Journal. 08/27/2008.

Interpretive Summary: Measuring imported fire ant populations on the ground is expensive and time consuming, and may not be possible in areas with limited access. ARS researchser have demonstrated the ability to detect up to 80% of fire ant mounds using remote sensing techniques, but manually counting mounds in aerial images is time consuming. More recent tools developed by ARS researchers in cooperation with an industry partner, enable automatic detection of mounds based upon unique shape and other mound characteristics. This accomplishment will make it easier and less expensive to process large amounts of aerial imagery for remote detection of fire ant mounds.

Technical Abstract: Imported fire ants construct earthen nests (=mounds) that exhibit many characteristics which make them potentially good targets for remote sensing programs, including geographical orientation, topography, and bare soil surrounded by actively growing vegetation. Template based features and object-based features extracted from aerial multispectral imagery of fire ant infested pastures were used to construct classifiers for automated fire ant mound detection. A classifier constructed using template based features alone yielded a 79% probability of detection with a corresponding false alarm rate of 9%. Addition of object based features (compactness and symmetry) to the classifier yielded a 79% probability of detection with a corresponding false alarm rate of 4%. Maintaining a 79% detection rate when applying the classifier to a second, unique pasture data set with different seasonal and other environmental factors resulted in a false alarm rate of 17.5%. Data demonstrate that automated detection of mounds with classifiers incorporating template and object based features is feasible, but it may be necessary to construct unique classifiers on a site-specific basis.